Computers do maths, not meanings. So every word is turned into a list of
numbers called an embedding. The clever bit: words that mean similar things get similar
numbers, so they land close together on a map. Click any word to explore.
The meaning map — click a word to find its neighbours
Tip: words near each other mean similar things. Try clicking puppy.
✨ Meaning-maths examples:
🤖 In a real LLM
Here each word is just 2 numbers (so we can draw it). A real model gives every token
hundreds of numbers — and it learns those numbers while training on tons of text.
Because "king" and "queen" end up with similar numbers, the model "knows" they're related —
without anyone ever telling it what a king is.